Search

Abstract Modelling of solar irradiation is paramount to renewable energy management. This warrants the inclusion of additive effects to predict solar irradiation. Modelling of additive effects to solar irradiation can improve the forecasting accuracy of prediction frameworks. To help develop the frameworks, this current study modelled the additive effects using non-parametric quantile regression (QR). The approach applies quantile splines to approximate non-parametric components when finding the best relationships between covariates and the response variable. However, some additive effects are perceived as linear. Thus, the study included the partial linearly additive quantile regression model (PLAQR) in the quest to find how best the additive effects can be modelled. As a result, a comparative investigation on the forecasting performances of the PLAQR, an additive quantile regression (AQR) model and the new quantile generalised additive model (QGAM) using out-of-sample and probabilistic forecasting metric evaluations was done. Forecasted density plots, Murphy diagrams and results from the Diebold–Mariano (DM) hypothesis test were also analysed. The density plot, the curves on the Murphy diagram and most metric scores computed for the QGAM were slightly better than for the PLAQR and AQR models. That is, even though the DM test indicates that the PLAQR and AQR models are less accurate than the QGAM, we could not conclude an outright greater forecasting performance of the QGAM than the PLAQR or AQR models. However, in situations of probabilistic forecasting metric preferences, each model can be prioritised to be applied to the metric where it performed slightly the best. The three models performed differently in different locations, but the location was not a significant factor in their performances. In contrast, forecasting horizon and sample size influenced model performance differently in the three additive models. The performance variations also depended on the metric being evaluated. Therefore, the study has established the best forecasting horizons and sample sizes for the different metrics. It was finally concluded that a 20% forecasting horizon and a minimum sample size of 10000 data points are ideal when modelling additive effects of solar irradiation using non-parametric QR.

This study proposes a more robust methodological approach to modeling the effect of weather and calendar variables on the number of bike rentals. We employ penalized splines quasi-Poisson regression (a semi-parametric model), which involves some form of regularization, like those used in lasso, ridge, and other types of parametric regularization models. We demonstrate that this modeling approach reveals hidden relationships that a pure parametric model fails to identify.The findings show that visibility, windspeed, season, working day, and year all significantly impact bike rentals. Increased rentals are associated with increased visibility and lower wind speed. Rentals are negatively affected by the spring and winter seasons, while working days and the year show positive trends except in a few cases. The analysis of rentals by registered and casual users reveals similar patterns, though the magnitudes of the effects differ. These findings highlight the importance of considering weather and calendar variables when managing and promoting bike-sharing services. The study has implications for bike-sharing system operators and policymakers, suggesting strategies such as improving visibility and wind protection, seasonally tailoring promotional campaigns, targeting non-working days for casual users, and adapting to changing user demands. The study adds to our understanding of the factors that influence bike rentals and provides suggestions for improving the utilization and accessibility of bike-sharing systems.

Dimensionality poses a challenge in developing quality predictive models. Often when modelling solar irradiance (SI), many covariates are considered. Training such data has several disadvantages. This study sought to identify the best variable embedded selection method for different location and time horizon combinations from Southern Africa solar irradiance data. It introduced new variable selection methods into solar irradiation studies, namely penalised quantile regression (PQR), regularised random forests (RRF), and quantile regression forest (QRF). Stability analysis, performance and accuracy metric evaluations were used to compare them with the common lasso, elastic and ridge regression methods. The QRF model performed best in all locations followed by the shrinkage methods on hourly data. However, it was found that QRF is not sensitive to associations through correlations, thereby ignoring the relevance of variables while focusing on importance. Among the shrinkage methods, the lasso performed best in only one location. On the 24-hour horizon, elastic net dominated the performances among the shrinkage methods, but QRF was best in three locations of the six considered. Results confirmed that variable selection methods performed differently on different situational data sets. Depending on the strengths of the methods, results were combined to identify the most paramount variables. Day, total rainfall, and wind direction were superfluous features in all situations. The study concluded that shrinkage methods are best in cases of extreme multicollinearity, while QRF is best on data sets with outliers or/and heavy tails.

Background: This study explores the determinants impacting the mortality risk of COVID-19 patients following hospitalisation within South Africa’s Limpopo province. Methods: Utilising a dataset comprising 388 patients, the investigation employs a frailty regression model to evaluate the influence of diverse characteristics on mortality outcomes, contrasting its performance against other parametric models based on loglikelihood measures. Results: The findings underscore diabetes and hypertension as notable contributors to heightened mortality rates, underscoring the urgency of effectively managing these comorbidities to optimise patient well-being. Additionally, regional discrepancies come to the fore, with the Capricorn district demonstrating elevated mortality risks, thereby accentuating the necessity for precisely targeted interventions. Medical interventions, particularly ventilation, emerge as pivotal factors in mitigating mortality risk. Gender-based distinctions in mortality patterns also underscore the need for bespoke patient care strategies. Conclusions: Collectively, these outcomes supply practical insights with implications for healthcare interventions, policy formulation, and clinical strategies aimed at ameliorating COVID-19 mortality risk among individuals discharged from hospitals within South Africa’s Limpopo province.

Financial time series data are known to exhibit volatility clustering, which implies that the volatility of financial returns tends to persist over time. This phenomenon has significant implications for risk management and financial decision-making. To capture this volatility clustering effect, the Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model is widely used. Through a Monte Carlo simulation experiment, the performance GARCH-type model with five different error distribution specifications were compared. The study found that the Skewed Generalized Error Distribution (SGED) was the most efficient and consistent distribution for modelling financial time series. The SGED outperformed the Normal, Student's t, GED, and skewed Student's t distributions in terms of goodness-of-fit, MSE, MAE and different sample sizes. The study suggests that the SGED may be more suitable for modelling financial time series with a GARCH-type specification in general contexts.

The S&P 500 Index is considered the most popular trading instrument in financial markets. With the rise of cryptocurrencies over the past few years, Bitcoin has grown in popularity and adoption. This study analyzes the daily return distribution of Bitcoin and the S&P 500 Index and assesses their tail probabilities using two financial risk measures. As a methodology, we use Bitcoin and S&P 500 Index daily return data to fit the seven-parameter General Tempered Stable (GTS) distribution using the advanced fast fractional Fourier transform (FRFT) scheme developed by combining the fast fractional Fourier transform algorithm and the 12-point composite Newton–Cotes rule. The findings show that peakedness is the main characteristic of the S&P 500 Index return distribution, whereas heavy-tailedness is the main characteristic of Bitcoin return distribution. The GTS distribution shows that 80.05% of S&P 500 returns are within −1.06% and 1.23% against only 40.32% of Bitcoin returns. At a risk level (α), the severity of the loss (AVaRα(X)) on the left side of the distribution is larger than the severity of the profit (AVaR1−α(X)) on the right side of the distribution. Compared to the S&P 500 Index, Bitcoin has 39.73% more prevalence to produce high daily returns (more than 1.23% or less than −1.06%). The severity analysis shows that, at α risk level, the average value-at-risk (AVaR(X)) of Bitcoin returns at one significant figure is four times larger than that of the S&P 500 Index returns at the same risk.

This study examined the biological, social, and clinical risk factors for mortality in coronavirus of the year 2019 (COVID-19) hospitalised patients. The population of the study is prone to COVID-19, thus understanding the most common traits and comorbidities of people who were affected is crucial in reducing its consequences. In this study, four supervised machine learning algorithms were implemented and compared to predict the mortality rate based on the explanatory variables across the five districts of Limpopo Province in South Africa. The data was obtained from Limpopo Department of Health. Prediction about the chances of dying from COVID-19 disease was made using logistic regression, random forest, support vector machine, and decision tree algorithms on the dataset of 20,592 records with twenty-one attributes. Due to the imbalanced nature of the data, Random Over-Sampling Examples (ROSE) were employed to balance our data for more accurate classification effectively. The ROSE package provides functions to deal with binary classification problems in the presence of imbalanced classes. We used 70% of the data for training, while 30% was selected for testing the predictive algorithms. A technique called Step Akaike's Information Criterion (StepAIC) was deployed to reduce the insignificant variables from the full model of the logistic regression. According to the findings of the study, among the four algorithms tested, random forest had the highest recall rate for predicting mortality at roughly 79 percent compared to the other three algorithms. Accordingly, we conclude that random forest algorithm is appropriate for predicting the chances of patients dying from COVID-19 based on the attributes of the five districts of Limpopo Province. In terms of the features and their importance, a function called Variable Importance (VarImp) was used to check which of the attributes have predictive power on the outcome variable (discharged status). The findings revealed that age, ever ventilated, ever oxygenated, intensive ward upon admission, Waterberg district, and private facility type were among the risk factors that could be selected for the logistic regression model to predict mortality in hospitalised patients. This implies that special attention should be given to these identified variables. The random forest model is adequate to establish these factors since the findings reveal that a fairly considerable percentage of explained variation would correctly classify 79% of the cases. The novelty of conducting research on the analysis of COVID-19 cases and comorbidities using machine learning algorithms lies in the potential to uncover new insights and patterns that might not be immediately apparent through traditional statistical methods. Machine learning algorithms can quickly and accurately analyze large datasets and identify complex relationships between variables, which could provide valuable information for public health officials and policymakers.

Background: South Africa has a high burden of tuberculosis (TB) disease and is currently not meeting the national and international reduction outcome targets. The TB prevalence rate of South Africa in 2015 was estimated at approximately 690 per 100,000 population per year, with an incidence rate of about 834 per 100,000 population. This study examines risk factors associated with development of TB in South Africa. Materials and Methods: This study utilised readily available open access secondary data of 2019 South African Health and Demographic Survey from Statistics South Africa (StatsSA) website, which was collected from self-reported information relating to TB in the household questionnaire. The factors analysed were of demographic, socio-economic and health nature. Bivariate and binary logistics analyses were carried out from which appropriate inferences were drawn on the association of TB with demographic, socio-economic and health factors. Results: In multivariate analysis the study revealed that age, personal weight, smoke, alcohol, asthma, province of residence, race and usually coughing were significantly associated with an increased risk of having TB. Conclusions and Recommendations: The results strongly suggest that young and older people coming from black and coloured ethic groups, who are asthmatic and cough frequently, and/or smoking and consuming alcohol are at high risk of developing TB. In addition, those who are overweight appear to have an increased risk of TB transmission, with the Western Cape, Eastern Cape, Northern Cape, Free State, North West and Gauteng being the hardest hit provinces. Hence, the study recommends that these factors must be taken into account in the planning and development of TB policies in order to work successfully towards the achievement of sustainable development goal of reducing TB by 80% before 2030.

Financial market data are abundant with outliers, and the search for an appropriate extreme value theory (EVT) approach to apply is an endless debate in the statistics of extremes research. This paper uses EVT methods to model the five-year daily all-share total return index (ALSTRI) and the daily United States dollar (USD) against the South African rand (ZAR) exchange rate of the Johannesburg stock exchange (JSE). The study compares the block maxima approach and the peaks-over-threshold (POT) approach in terms of their ability to model financial market data. The 100-year return levels for the block maxima approach were found to be almost equal to the maximum observations of the financial markets of 10,860 and R18.99 for the ALSTRI and the USD–ZAR, respectively. For the peaks-over-threshold (POT) approach, the results show that the ALSTRI and the USD–ZAR exchange rate will surpass 17,501.63 and R23.72, respectively, at least once in 100 years. The findings in this study reveal a clear distinction between block maxima and POT return level estimates. The POT approach return level estimates were comparably higher than the block maxima estimates. The study further revealed that the blended generalised extreme value (bGEVD) is more suitable for relatively short-term forecasting, since it cuts off at the 50-year return level. Therefore, this study will add value to the literature and knowledge of statistics and econometrics. In the future, more studies on bGEVD, vine copulas, and the r-largest-order bGEVD can be conducted in the financial markets.

This study proposes a modified Geometric Brownian motion (GBM), to simulate stock price paths under normal and convoluted distributional assumptions. This study utilised four selected continuous probability distributions for the convolution because of shared properties, including normality, and parameters that have a standard distribution with a location and scale parameters of zero and one, in that order. The findings from this study revealed that the simulation of price paths looks identical under the assumption of normal distribution and normal convolved with normal, Laplace, and Rice distributions for different sample sizes and parameter settings but differs with respect to the Cauchy distribution. Furthermore, the study found that all the mean absolute percentage error (MAPE) and mean square error (MSE) values for the normal and convoluted distributions underlying the GBM were approximately less than 10%, indicating high forecast accuracy. However, the average simulated price paths for the GBM under the normal distribution was found to be significantly different from the GBM under convoluted distribution when a t-test was employed for different sample sizes and different settings of the drift and volatility values.

Exploration of solar irradiance can greatly assist in understanding how renewable energy can be better harnessed. It helps in establishing the solar irradiance climate in a particular region for effective and efficient harvesting of solar energy. Understanding the climate provides planners, designers and investors in the solar power generation sector with critical information. However, a detailed exploration of these climatic characteristics has not yet been studied for the Southern African data. Very little exploration is being done through the use of measures of centrality only. These descriptive statistics may be misleading. As a result, we overcome limitations in the currently used deterministic models through the application of distributional modelling through quantile functions. Deterministic and stochastic elements in the data were combined and analysed simultaneously when fitting quantile distributional function models. The fitted models were then used to find population means as explorative parameters that consist of both deterministic and stochastic properties of the data. The application of QFs has been shown to be a practical tool and gives more information than approaches that focus separately on either measures of central tendency or empirical distributions. Seasonal effects were detected in the data from the whole region and can be attributed to the cyclical behaviour exhibited. Daily maximum solar irradiation is taking place within two hours of midday and monthly accumulates in summer months. Windhoek is receiving the best daily total mean, while the maximum monthly accumulated total mean is taking place in Durban. Developing separate solar irradiation models for summer and winter is highly recommended. Though robust and rigorous, quantile distributional function modelling enables exploration and understanding of all components of the behaviour of the data being studied. Therefore, a starting base for understanding Southern Africa’s solar climate was developed in this study.

Extreme rainfall events have made significant damages to properties, public infrastructure and agriculture in some provinces of South Africa notably in KwaZulu-Natal and Gauteng among others. The general global increase in the frequency and intensity of extreme precipitation events in recent years is raising a concern that human activities might be heavily disturbed. This study attempts to model long-term monthly rainfall variability in the selected provinces of South Africa using various statistical techniques. The study investigates the normality and stationarity of the underlying distribution of the whole body of rainfall data for each selected province, the long-term trends of the rainfall data and the extreme value distributions which model the tails of the rainfall distribution data. These approaches were meant to help achieve the broader purpose of this study of investigating the long-term rainfall trends, stationarity of the rainfall distributions and extreme value distributions of monthly rainfall records in the selected provinces of South Africa in this era of climate change. The five provinces considered in this study are Eastern Cape, Gauteng, KwaZulu-Natal, Limpopo and Mpumalanga. The findings revealed that the long-term rainfall distribution for all the selected provinces does not come from a normal distribution. Furthermore, the monthly rainfall data distribution for the majority of the provinces is not stationary. The paper discusses the modelling of monthly rainfall extremes using the non-stationary generalised extreme value distribution (GEVD) which falls under the block maxima extreme value theory (EVT) approach. The maximum likelihood estimation method was used to obtain the estimates of the parameters. The stationary GEVD was found as the best distribution model for Eastern Cape, Gauteng, and KwaZulu-Natal provinces. Furthermore, model fitting supported non-stationary GEVD model for maximum monthly rainfall with nonlinear quadratic trend in the location parameter and a linear trend in the scale parameter for Limpopo, while in Mpumalanga the non-stationary GEVD model with a nonlinear quadratic trend in the scale parameter and no variation in the location parameter fitted well to the monthly rainfall data. The negative values of the shape parameters for Eastern Cape and Mpumalanga suggest that the data follow the Weibull distribution class, while the positive values of the shape parameters for Gauteng, KwaZulu-Natal and Limpopo suggest that the data follow the Fréchet distribution class. The findings from this paper could give information that can assist decision makers establish strategies for proper planning of agriculture, infrastructure, drainage system and other water resource applications in the South African provinces.

In this article we fit a time-dependent generalised extreme value (GEV) distribution to annual maximum flood heights at three sites: Chokwe, Sicacate and Combomune in the lower Limpopo River basin of Mozambique. A GEV distribution is fitted to six annual maximum time series models at each site, namely: annual daily maximum (AM1), annual 2-day maximum (AM2), annual 5-day maximum (AM5), annual 7-day maximum (AM7), annual 10-day maximum (AM10) and annual 30-day maximum (AM30). Non-stationary time-dependent GEV models with a linear trend in location and scale parameters are considered in this study. The results show lack of sufficient evidence to indicate a linear trend in the location parameter at all three sites. On the other hand, the findings in this study reveal strong evidence of the existence of a linear trend in the scale parameter at Combomune and Sicacate, whilst the scale parameter had no significant linear trend at Chokwe. Further investigation in this study also reveals that the location parameter at Sicacate can be modelled by a nonlinear quadratic trend; however, the complexity of the overall model is not worthwhile in fit over a time-homogeneous model. This study shows the importance of extending the time-homogeneous GEV model to incorporate climate change factors such as trend in the lower Limpopo River basin, particularly in this era of global warming and a changing climate. Keywords: nonstationary extremes; annual maxima; lower Limpopo River; generalised extreme value

Short-term hourly load forecasting in South Africa using additive quantile regression (AQR) models is discussed in this study. The modelling approach allows for easy interpretability and accounting for residual autocorrelation in the joint modelling of hourly electricity data. A comparative analysis is done using generalised additive models (GAMs). In both modelling frameworks, variable selection is done using least absolute shrinkage and selection operator (Lasso) via hierarchical interactions. Four models considered are GAMs and AQR models with and without interactions, respectively. The AQR model with pairwise interactions was found to be the best fitting model. The forecasts from the four models were then combined using an algorithm based on the pinball loss (convex combination model) and also using quantile regression averaging (QRA). The AQR model with interactions was then compared with the convex combination and QRA models and the QRA model gave the most accurate forecasts. Except for the AQR model with interactions, the other two models (convex combination model and QRA model) gave prediction interval coverage probabilities that were valid for the 90 % , 95 % and the 99 % prediction intervals. The QRA model had the smallest prediction interval normalised average width and prediction interval normalised average deviation. The modelling framework discussed in this paper has established that going beyond summary performance statistics in forecasting has merit as it gives more insight into the developed forecasting models.